agent: builder model: opus max_turns: 100 budget_usd: 20 —
if or match is handed back (#1085)Score: blocks #928 step 4 (yaml’s events, ADR-yaml-1 D7; y21-events run 3 blocked on it, draft in local branch wip/y21-events at 1c1199d6). A compiler defect found by the yaml lane; fixed in this lane because only yaml needs it now. It changes crates/: name in the report what it decided (the supervisor raises the version), Changelog: one sentence. The commit says Fixes #1085.
Facts: Where things live, Traps already met
A value holding views of a ref parameter handed back through an if or match expression is refused NK2302 (“many keeps n after this call, but n: ref N doesn’t say whose data it points into”); handed back directly (return [E { name: n.name }], return E { .. }) it passes since #1071 (79ee176c). It shows only where the struct’s field is a view: below, parse builds N from a trim(). A struct N built in main from a literal does not show it (the field is then owned), which is why y21’s own probe passed for the architect.
target/fast/nikaia run a.nika; today NK2302; wanted 1 hello; checked by the architect at 66f7750f)struct N {
name: String,
}
struct E {
name: String,
}
fn many(n: ref N) -> Vec[E] sync {
return if true { [E { name: n.name }] } else { [] }
}
fn all(ns: ref Vec[N]) -> Vec[E] sync {
let mut out: Vec[E] = []
for n in ns {
out.extend(many(n))
}
return out
}
fn parse(input: String) -> Vec[N] {
return [N { name: input.trim() }]
}
fn main() {
let ns = parse(" hello ")
let es = all(ns)
println(f"{es.len()} {es[0].name}")
}Variants, each wanted 1 hello with the same all/parse/main (all refused today):
many: return match 1 { / 1 => [E { name: n.name }], / else => [], / }many: let r: Vec[E] = if true { [E { name: n.name }] } else { [] } then return rmany: an arm that is a block, return if true { let mut out: Vec[E] = [E { name: n.name }] / out } else { [] } (yaml’s walk arms are such blocks, with for/extend/push inside)fn one(n: ref N) -> E sync { return if true { E { name: n.name } } else { E { name: n.name } } }, with all doing out.push(one(n))crates/nikaia-std/src/tools/views.nika, handed_back (~1076): a StructLit of the result and a ListLit (#1071) are handed back, everything else is stores. An Expr::If, Expr::Match, Expr::Block is to be followed to the value each branch or arm ends in (a block’s tail, an arm’s expression), each asked as handed_back; the condition and the matched value are asked as today (descend/stores on the parts that are not handed back). Look at how rsrc_if/rsrc_match/rsrc_block (~2185) find the tails: reuse their shape.let r = <if/match>; return r shape: local_vec_handed_back / struct_handed_back (~880, ~983) look at the let’s value; they need the same descent.let mut out = [..] inside an arm and its out tail is #1071’s local-list case one level down; check that local_vec_handed_back sees the arm’s block, not only the function’s body.tools/*.nika change, lower-std regenerates its lowered/*.rs; commit both)handed_back follows if/match/block tails: the probe and its match, block-arm and struct-result variants print 1 hello. Test tests/compiler/tests/views_handed_back_through_if_and_match.nika with .stdout (all four shapes in one program).let r = <if/match>; return r shape the same, in that test. A branch that stores the view elsewhere stays refused: tests/compiler/tests/view_stored_in_a_branch_refused.nika with .stderr (e.g. an arm that assigns n.name to a field of a self whose struct does not carry n’s buffer). tests/compiler --both-settings green; the #1071 tests (rg -l 1071 tests/compiler/tests) unchanged.git show 1c1199d6:packages/yaml/src/main.nika > /tmp/y.nika (scratchpad path of your own), target/fast/nikaia check on it shows no error (the architect saw one NK2302 at walk’s return match n.value; with walk rewritten to push into a local the rest of the draft builds and its events test runs). Say in Tests: that it checks. Do not commit anything under packages/yaml/: y21 run 4 takes the draft.Files you may change: the tools/*.nika under Locations with their lowered/*.rs, the new tests named under To build, and an existing test’s expected output your change rightly changes (name it in Body:). If the fix needs a language decision, stop: Result: blocked, Needs: architect decision. Stop about 15 turns before max_turns, commit what is finished, report partial.
docs/builder-facts.md first. develop in the main checkout; git status at start: other agents’ uncommitted files are not yours, never touch or stage them.git add <files> only; commit after each step. Do not push: the driver runs the suite on your commit and pushes it (CLAUDE.md, Builders commit, the driver tests and pushes).cargo fmt, cargo clippy --all-targets -- -D warnings, scripts/pre-push-check.sh, and the gate of .claude/agents/builder.md (#852).CHANGELOG.md, badges; no new NK code. New tests in Nikaia only; never the full suite. No foreground sleep, no polling; scripts in files.std or the toolchain lacks stops the work: no code around it unless an ADR or Part I-III refuses it (name it). Else Result: blocked, Found: gap + probe, Needs: architect decision (CLAUDE.md, A missing form stops the work).NIKAIA_CACHE_LIMIT=2G; at most two clones with target/ in the machine; delete your own clone’s target/ and target/release/nikaia-scratch at the end; report GB freed.Pushed: line is local <sha> of your last commit. A report without a Pushed: line, or with a placeholder left in (TC, DISK), is not done and cuts the batch (k3a1 2026-10-29: Result: done, nothing committed).NIKAIA_CACHE_LIMIT=2G target/fast/nikaia test --project packages/yaml green, and cargo test --profile fast -p nikaia --test it -- fmt::every_nika_file (it formats every .nika under packages/). For a change only under packages/yaml/ this is the gate the line above asks for; a change under crates/ runs the gate of .claude/agents/builder.md (#852) as well. target/fast/nikaia is built by the session hook; if it is missing, cargo build --profile fast -p nikaia once.Changelog: none, unless the brief changes crates/.Files not to touch: anything outside packages/yaml/ and the brief’s own report, unless the brief names it; never docs/briefs/batch.txt or a brief of another lane.
max_turns: commit every finished step, then report Result: partial with what is left under Needs:. Uncommitted work at the end is lost work (y4 ran out at 110)./home/user/Nikaia/docs/briefs/yaml/y28-1085-views-back-through-if-match.md.report.md)Fields of docs/architect.md, Briefs and reports: Result: done|noop|partial|blocked (the bare word alone on that line: Result: done, never a summary after it, which cuts the batch, b869a 2026-10-27; Owner: is a question for the owner or none, never a role), Files, Commit, Body (optional, one paragraph for the commit message), Fixes, Pushed, Tests, Gate (green|red), Deviation, Found, Needs, Changelog, Owner, Disk freed, and for a move Rs2nika: (the kept N of M lines of rs2nika --score, then the three costliest hand edits; none for a task that moves no Rust, none (redesign) for a brief with Shape: redesign and no Convert: line). none where nothing, and then the bare word: Deviation: none (a note) cuts the batch (batch.nika compares exactly); a note goes into the commit message. Found: is only for a defect outside your change (it cuts the batch); a process event such as another session resetting or rewriting your files is no Found:, it goes under Needs: (m3a 2026-10-28); a converter draft that came out wrong is no Found:, it is one of the costliest edits under Rs2nika:, which the converter lane reads (e12-4b 2026-10-27); how you solved your own step goes into the commit message. Deviation: is only a step of the brief built otherwise than written or not built (a choice the brief left to you, such as keep what X needs and say so, is no deviation: it goes into the commit message; nor is building two steps in one commit when every step is built, e12-3a 2026-10-27, nor a step another commit on develop already built, shown by your test failing with that commit’s change reverted, b889 2026-10-27: name the commit in Body:, nor a test’s expected .stdout/.stderr taken from what the program really prints where the brief’s sketch of it was short or wrong, b934 2026-10-29: say so in Body:, nor a lowering that follows the ADR governing it where the brief’s expected lowering contradicts that ADR, b933 2026-10-29: name the ADR in Body:, nor a lowering other than the brief’s where rustc cannot take the brief’s, b1029 2026-10-31: say why in Body:, nor re-keying an allow-list entry (KNOWN) whose hash changed for the same program and the same reason, b-fixforward-sweep2 2026-10-28: say so in Body:, nor an existing test’s expected .stdout/.stderr updated because your change rightly changes the message it prints, b1012b 2026-10-31: name the test in Body:); how you verified (e.g. a test not run against the unfixed compiler) goes under Tests:. A noop says why under Tests: (what was probed), not under Deviation:. Final answer, one line: done <sha>, report <path> (or BLOCKED: ... / PUSH REJECTED ...).
Attribution: Co-Authored-By: Claude Opus 5.5 noreply@anthropic.com Claude-Session: https://claude.ai/code/session_01UVxJDveRM5qKZKyCqq6Y24